Threat analysis of video surveillance system and security solutions

Abstract

Nowadays, video surveillance systems are widely used by companies and private customers to enhance the security of the desired area. Modern video surveillance systems can offer continuous surveillance and provide a diverse set of features for maintaining security and incident investigation at a high level. Such features include playback, fast-forwarding video functions; face, car plates, and object recognition; alarms notifying about suspicious movement detection on the video camera, and many others. These features are supported by such services as artificial intelligence servers and web applications. However, new technologies and services that create these features cause new vulnerabilities and threats in the system. This thesis describes video surveillance system architecture and confidential information data flow in the system, explores threats, and system assets, proposes methods for risk assessment and threat prioritization for the video surveillance system's services, and presents a risk assessment on the example of a web application. The main contribution is the extensive threat analysis of the system components following the STRIDE model and risk assessment based on the identified threats. Furthermore, this work proposes several technical solutions for video surveillance system security: physical and information security, privacy protection, access right violation prevention, and system reliability support.